VC
Value Add VC
โšกHomePulseโšกHelpful Apps๐Ÿ“Blog๐ŸคPartner
Illustration for: Nvidia Competitor Etched Hits $5B Valuation on $1B in AI Chip Sales
Value Add VC/Pulse/FUNDING$5B valuation

Nvidia Competitor Etched Hits $5B Valuation on $1B in AI Chip Sales

Etched, the AI inference chip startup, confirmed a $5 billion post-money valuation on June 30 following a $500 million round led by Stripes, bringing total funding to $800 million since its 2022 founding. The company says it has already secured $1 billion in contract orders for its 'frontier inference clusters' after TSMC successfully manufactured its custom chips earlier in 2026, with backers including Jane Street, Two Sigma, Andrej Karpathy, Geoffrey Hinton, Fei-Fei Li, Stanley Druckenmiller and Peter Thiel.

By the Numbers

$5B
Post-Money Valuation
$500M, led by Stripes, Dec 2025
Latest Round
$800M
Total Raised Since 2022
$1B (frontier inference clusters)
Contract Orders
TSMC-fabricated chips, early 2026
Manufacturing Milestone
NvidiaEtched
TC
By the Funding Desk
Edited by Trace Cohen ยท Early-stage VC & angel ยท Founder, New York Venture Partners
June 30, 2026
3 min read
ShareXLinkedInEmail

THE RUNDOWN

1

$1B in contract orders before broad commercial availability is a rare, concrete demand signal for a chip startup still primarily known for its funding rounds rather than shipped volume

2

Successful TSMC manufacturing of Etched's custom silicon in 2026 clears one of the hardest technical and supply-chain milestones any chip startup faces โ€” actual fabrication at scale

3

An angel list spanning Karpathy, Hinton, Fei-Fei Li, Druckenmiller and Thiel signals unusually broad conviction across both AI research luminaries and traditional finance, not just typical deep-tech VCs

4

Directly ties into this week's Thiel Capital/Copper Sky story, illustrating how thoroughly the Thiel investment network runs through Etched's cap table from multiple angles

TC

The VC Read ยท Trace's Take

Trace Cohen

A chip startup landing $1B in contract orders after actually clearing TSMC fabrication โ€” not just raising a big round on a promising roadmap โ€” is the real signal here; most AI hardware startups die at exactly the manufacturing-scale step Etched just cleared. The investor list doing double duty as both a credibility signal (Karpathy, Hinton, Fei-Fei Li actually vouching for the technical bet) and a capital signal (Druckenmiller, Thiel, quant trading firms) is about as strong a combination as a chip startup can assemble at this stage. The Thiel-network overlap with Copper Sky's Series A stake is worth watching as a pattern โ€” this is now at least the second AI infrastructure win traced back to the same investor ecosystem this week alone. For founders in AI hardware, Etched proves specialized inference silicon can win real orders against Nvidia's dominance if the fabrication and performance claims actually hold up under customer testing. Watch whether that $1B order book converts into recognized revenue at anything close to the pace implied โ€” that's the number that separates a real business from a very well-funded chip design.

โš”๏ธ AI Chip Wars โ†’๐Ÿ’พ AI Chip Startups โ†’

Analysis

Etched, the AI inference chip startup founded by Harvard dropouts and Thiel Fellows Gavin Uberti and Robert Wachen, confirmed a $5 billion post-money valuation on June 30, 2026, following a $500 million funding round completed in December 2025 and led by investment firm Stripes. The round brings Etched's total funding to $800 million since its 2022 founding, and comes alongside a genuinely concrete commercial milestone: $1 billion in contract orders for its 'frontier inference clusters' โ€” complete systems combining Etched's custom chips, server racks and software designed specifically to accelerate AI model inference.

The technical validation underpinning the valuation is significant on its own: TSMC successfully manufactured Etched's custom chips earlier in 2026, clearing one of the hardest milestones any AI chip startup faces, well before most competitors reach commercial-scale fabrication. The company is now testing its inference clusters directly with customers, positioning the product as faster, more efficient and lower-cost than general-purpose GPU-based inference for frontier AI models specifically.

Etched's investor base spans an unusually wide range of both AI-research credibility and traditional finance conviction. Beyond lead investor Stripes, backers include quantitative trading firms Jane Street, Hudson River Trading and Two Sigma, along with VentureTech Alliance and Ribbit Capital. Angel investors include AI researchers Andrej Karpathy, Geoffrey Hinton and Fei-Fei Li โ€” three of the field's most closely watched names โ€” alongside billionaire investors Stanley Druckenmiller and Peter Thiel, giving the company backing from both the technical research community most qualified to judge the chip's actual capabilities and financiers most focused on its commercial trajectory.

โ€œEtched's investor base spans an unusually wide range of both AI-research credibility and traditional finance conviction.โ€

The Thiel connection runs deeper than a single angel check: Copper Sky Capital, the Phoenix-based firm run by Thiel Capital managing director Jack Selby, separately invested in Etched's roughly $120 million Series A about two years ago, partly on the strength of a pledge to help move Etched's chip fabrication toward Arizona's TSMC facility โ€” meaning Etched's cap table now reflects Thiel-network involvement from at least two distinct angles, an early institutional check and a personal angel investment, both converging on the same company's now-validated $5 billion outcome.

Etched's specific bet is that specialized, application-specific inference chips can outperform general-purpose GPUs for the narrow but enormous task of running already-trained frontier models efficiently โ€” a different wager than Nvidia's broad, flexible GPU architecture, and one that requires genuinely novel chip design rather than incremental improvement. The $1 billion order book, if it converts into delivered, paid revenue at anything close to that scale, would represent unusually fast commercial traction for a chip company at this stage.

For founders and investors in AI infrastructure and chips, Etched's trajectory โ€” successful custom fabrication, a broad and credible investor base, and real contracted demand โ€” is a genuine proof point that application-specific inference silicon can attract both capital and customers at scale, even against Nvidia's dominant market position. For LPs, the diversity of Etched's cap table (quant trading firms, traditional VC, AI research luminaries, and Thiel-network capital from multiple vehicles) is a useful illustration of how broadly capital has organized around a single credible chip bet.

What to watch: whether Etched's $1 billion in contract orders converts into delivered, recognized revenue at the pace implied, how the company's inference clusters perform in real customer deployments relative to Nvidia GPU-based alternatives, and whether Etched pursues a public listing or additional private rounds as it scales manufacturing further.

Related Deep Dives

  • Hadrian's $1.37B Series D: Defense Manufacturing Hits an ... โ†’
  • World Labs Valuation: What Fei-Fei Li's Spatial Intellige... โ†’
  • $3.5M ARR, 120% NRR โ€” Series A AI Bar (2026) โ†’
ShareXLinkedInEmail

More on

Nvidia โ†’Etched โ†’

Prior Pulse Coverage

NvidiaWhy the AI Buildout Runs on Debt, Not VCsNvidiaOpen-Weight Labs Become the Valley's Acquisition TargetNvidiaNvidia Pauses Its Own AI Cloud Financing PlanNvidiaNvidia Now Carries $366B in Future CommitmentsNvidiaNvidia Is Financing Its Own Demand Curve

Key Sources

2 sources
SourceTechCrunch
AnalysisValue Add Pulse

Reported by TechCrunch ยท Analysis by Value Add Pulse.

โ† Back to Pulse

THE WIRE in your inboxโ€” Tech, startup & VC news with Trace's take. Free, no spam.

Read Next

FUNDINGยท Aug 29, 2026

Why the AI Buildout Runs on Debt, Not VCs

Illustration for: Why the AI Buildout Runs on Debt, Not VCs
FUNDING

Why the AI Buildout Runs on Debt, Not VCs

Nvidia's pause on financing smaller AI clouds, alongside a16z's new $1.1B hardware fund, shows two separate capital markets now underwriting AI: equity chasing the model layer, debt absorbing the risk of the physical buildout.

FUNDINGยท Aug 28, 2026

a16z Raises $1.1B to Build AI's Physical Layer

Illustration for: a16z Raises $1.1B to Build AI's Physical Layer
FUNDING$1.1B fund

a16z Raises $1.1B to Build AI's Physical Layer

Andreessen Horowitz has closed a $1.1 billion Machine Age fund dedicated to chips, memory, cooling, power and robotics -- the hardware underneath AI rather than the software on top of it.

FUNDINGยท Aug 28, 2026

Owner Raises $240M at $2.3B to Run Local Businesses

Illustration for: Owner Raises $240M at $2.3B to Run Local Businesses
FUNDING$240M at $2.3B

Owner Raises $240M at $2.3B to Run Local Businesses

Owner, the AI platform that runs websites, ordering, marketing and phone support for restaurants, raised $240 million led by Growth Equity at Goldman Sachs Alternatives at a $2.3 billion valuation.

Deep Dives

Hadrian's $1.37B Series D: Defense Manufacturing Hits an ...World Labs Valuation: What Fei-Fei Li's Spatial Intellige...$3.5M ARR, 120% NRR โ€” Series A AI Bar (2026)
@Trace_Cohenยทt@nyvp.com